The Role of Acute Toxicity Tests with Aquatic Animals in Chemical Registration and Notification Schemes

Abstract

Existing and proposed chemical registration and notification schemes include a requirement for acute toxicity test data on representative aquatic vertebrate and invertebrate species. These data, taken in conjunction with the pattern of disposal and use of the chemical, its physical and chemical properties, its stability and its potential for biodegradation and transport in the aquatic ecosystem, together with data on mammalian toxicity, contribute towards the initial hazard evaluation. A decision is then made, inter alia, whether further aquatic toxicity tests should be requested. In order to keep such requests to a minimum, it is essential that the maximum amount of useful information should be extracted from the basic toxicity tests. There has been a recent tendency in standard fish toxicity tests to report only the 96-h LC50 together with confidence limits; the statistical treatment of mortality data has been developed to ensure the maximum accuracy of the derived LC50. Such superficial refinements tend to obscure the scope for inaccuracies in these simple biological tests. The data which are of much greater value are a time series of LC50's within the 96-h period (3, 6, 24, 48, 72 h) from which a toxicity curve can be constructed. These curves have been recognised for many years as providing valuable information on the toxicity of chemicals and form a sound foundation on which to judge the validity of the test technique used for a particular chemical, and if necessary the type of further tests required. Examples will be given of various types of curves and their possible interpretations. An essential requisite for obtaining accurate toxicity curves is that the test concentrations of a chemical must be kept as constant as possible. Because of the problems and expense of chemical analyses to check whether concentrations of a substance have remained close to the nominal value in static toxicity tests, the basic test technique should be flow-through; static tests can be used only if the existing chemical, physical and biological data indicate that there will be no significant loss of toxicant from the test solution.